A General Tracking Control Framework for Uncertain Systems With Exponential Convergence Performance

Abstract

A difficult problem of exponential tracking control for uncertain systems even with external disturbance is investigated. The systems studied have a general/representative form with the lossless second-order differential systems and Euler-Lagrange systems included. A general control design framework is presented using observer technique. An observer-based estimator is first developed to precisely estimate and or reconstruct the uncertainty and the disturbance, and an estimator-based controller is then developed. Globally exponential stability of the closed-loop tracking system can be achieved by that controller. Moreover, the resulted estimation error of uncertainty can be stabilized with finite-time convergence. The key advantage of this control architecture is that the controller is able to achieve a perfect tracking performance with no overshoot observed and the settling time tuned to be as less as possible. The effectiveness of the approach is validated on a rigid-flexible coupling satellite example.